Objective¶

The goal of this analysis is to understand the demographic and usage patterns of Netflix users, calculate key metrics such as Lifetime Value (LTV), and provide insights based on various attributes such as age, country, device usage, and subscription types.

Data¶

The dataset contains the following columns:

• User ID: A unique identifier for each user.

• Subscription Type: The type of subscription plan (Basic, Standard, Premium).

• Monthly Revenue: The monthly revenue generated from each user.

• Join Date: The date the user joined Netflix.

• Last Payment Date: The date of the last payment made by the user.

• Country: The country of the user.

• Age: The age of the user.

• Gender: The gender of the user.

• Device: The primary device used by the user (Smartphone, Tablet, Smart TV, Laptop).

• Plan Duration: The duration of the subscription plan.

Importing Libraries¶

In [26]:
import pandas as pd
import seaborn as sns
import matplotlib.pyplot as plt
import math

Importing Dataset¶

In [2]:
df = pd.read_csv('Netflix Userbase.csv')
In [3]:
df.head()
Out[3]:
User ID Subscription Type Monthly Revenue Join Date Last Payment Date Country Age Gender Device Plan Duration
0 1 Basic 10 15-01-22 10-06-23 United States 28 Male Smartphone 1 Month
1 2 Premium 15 05-09-21 22-06-23 Canada 35 Female Tablet 1 Month
2 3 Standard 12 28-02-23 27-06-23 United Kingdom 42 Male Smart TV 1 Month
3 4 Standard 12 10-07-22 26-06-23 Australia 51 Female Laptop 1 Month
4 5 Basic 10 01-05-23 28-06-23 Germany 33 Male Smartphone 1 Month
In [4]:
# Checking the duplicates

df.duplicated().sum()
Out[4]:
0
In [5]:
# Checkinh the null values

df.isnull().sum()
Out[5]:
User ID              0
Subscription Type    0
Monthly Revenue      0
Join Date            0
Last Payment Date    0
Country              0
Age                  0
Gender               0
Device               0
Plan Duration        0
dtype: int64
In [6]:
df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 2500 entries, 0 to 2499
Data columns (total 10 columns):
 #   Column             Non-Null Count  Dtype 
---  ------             --------------  ----- 
 0   User ID            2500 non-null   int64 
 1   Subscription Type  2500 non-null   object
 2   Monthly Revenue    2500 non-null   int64 
 3   Join Date          2500 non-null   object
 4   Last Payment Date  2500 non-null   object
 5   Country            2500 non-null   object
 6   Age                2500 non-null   int64 
 7   Gender             2500 non-null   object
 8   Device             2500 non-null   object
 9   Plan Duration      2500 non-null   object
dtypes: int64(3), object(7)
memory usage: 195.4+ KB

Finding Number of users per Country¶

In [7]:
# Bar Chart 

plt.figure(figsize=(12,6))
df['Country'].value_counts().plot(kind='bar' , color='skyblue')
plt.title('Number of Users per Country')
plt.xlabel('Country')
plt.ylabel('Number of users')
plt.xticks(rotation=45)
plt.show()
No description has been provided for this image

Loop chart acc to Number of users¶

In [8]:
columns_titles = {
    'Gender': 'Gender Counts',
    'Device': 'Device Counts',
    'Subscription Type': 'Subscription type Counts'
    
}

for column, title in columns_titles.items():
    plt.figure(figsize=(12,6))
    df[column].value_counts().plot(kind='bar' , color='skyblue')
    plt.title(title)
    plt.xlabel(column)
    plt.ylabel('Count')
    plt.xticks(rotation=45)
    plt.show()
No description has been provided for this image
No description has been provided for this image
No description has been provided for this image

Map View¶

In [9]:
user_count_by_country = df['Country'].value_counts().reset_index()
user_count_by_country.columns = ['Country','User Count']

import plotly.express as px
In [10]:
fig = px.choropleth(user_count_by_country,
                    locations='Country',
                    locationmode='country names', 
                    color='User Count',
                    hover_name='Country',
                    title='User Distribution by Country',
                    color_continuous_scale= px.colors.sequential.Plasma
                   )

fig.update_layout(
     width = 1000,
     height = 700,
     title_font_size = 24,
     geo = dict(
     showframe = False,
     showcoastlines = False)
)

fig.show()

Subscription by Country¶

In [11]:
plt.figure(figsize=(12,6))
sns.countplot(x='Country', hue='Subscription Type',
             data = df, palette='Set2')
plt.title('Subscription Type by Country')
plt.xticks(rotation=45)
plt.show()
No description has been provided for this image
In [12]:
subs_by_country = df.pivot_table(index='Country',
                                columns='Subscription Type',
                                values='User ID',
                                aggfunc='count',
                                fill_value=0)

subs_by_country
Out[12]:
Subscription Type Basic Premium Standard
Country
Australia 31 101 51
Brazil 146 33 4
Canada 145 88 84
France 36 147 0
Germany 149 3 31
Italy 176 4 3
Mexico 4 0 179
Spain 110 212 129
United Kingdom 3 0 180
United States 199 145 107
In [13]:
subs_by_country['Total']= subs_by_country.sum(axis=1)
   
subs_by_country
Out[13]:
Subscription Type Basic Premium Standard Total
Country
Australia 31 101 51 183
Brazil 146 33 4 183
Canada 145 88 84 317
France 36 147 0 183
Germany 149 3 31 183
Italy 176 4 3 183
Mexico 4 0 179 183
Spain 110 212 129 451
United Kingdom 3 0 180 183
United States 199 145 107 451

Gender distribution by Country¶

In [14]:
plt.figure(figsize=(12,6))
sns.countplot(x='Country', hue='Gender',
             data = df, palette='pastel')
plt.title('Gender distribution by Country')
plt.xticks(rotation=45)
plt.show()
No description has been provided for this image
In [15]:
gender_by_country = df.pivot_table(index='Country',
                                columns='Gender',
                                values='User ID',
                                aggfunc='count',
                                fill_value=0)

gender_by_country
Out[15]:
Gender Female Male
Country
Australia 89 94
Brazil 95 88
Canada 157 160
France 91 92
Germany 94 89
Italy 91 92
Mexico 89 94
Spain 233 218
United Kingdom 93 90
United States 225 226
In [16]:
subs_by_country_gender= df.pivot_table(index='Country',
                                      columns=['Subscription Type','Gender'],
                                      values='User ID',
                                      aggfunc='count',
                                      fill_value=0)

subs_by_country_gender
Out[16]:
Subscription Type Basic Premium Standard
Gender Female Male Female Male Female Male
Country
Australia 16 15 45 56 28 23
Brazil 78 68 16 17 1 3
Canada 68 77 40 48 49 35
France 17 19 74 73 0 0
Germany 80 69 1 2 13 18
Italy 86 90 3 1 2 1
Mexico 2 2 0 0 87 92
Spain 60 50 113 99 60 69
United Kingdom 2 1 0 0 91 89
United States 103 96 72 73 50 57
In [17]:
subs_by_country_gender2= df.pivot_table(index=['Country','Gender'],
                                      columns='Subscription Type',
                                      values='User ID',
                                      aggfunc='count',
                                      fill_value=0)

subs_by_country_gender2
Out[17]:
Subscription Type Basic Premium Standard
Country Gender
Australia Female 16 45 28
Male 15 56 23
Brazil Female 78 16 1
Male 68 17 3
Canada Female 68 40 49
Male 77 48 35
France Female 17 74 0
Male 19 73 0
Germany Female 80 1 13
Male 69 2 18
Italy Female 86 3 2
Male 90 1 1
Mexico Female 2 0 87
Male 2 0 92
Spain Female 60 113 60
Male 50 99 69
United Kingdom Female 2 0 91
Male 1 0 89
United States Female 103 72 50
Male 96 73 57

Users by Age¶

In [18]:
plt.figure(figsize=(12,6))
df['Age'].value_counts().sort_index().plot(kind='bar',
                                           color='blue',
                                          edgecolor='black')


plt.title('Age Count')
plt.xlabel('Age')
plt.ylabel('Count')

plt.show()
No description has been provided for this image

Subscription Duration¶

In [19]:
df.head()
Out[19]:
User ID Subscription Type Monthly Revenue Join Date Last Payment Date Country Age Gender Device Plan Duration
0 1 Basic 10 15-01-22 10-06-23 United States 28 Male Smartphone 1 Month
1 2 Premium 15 05-09-21 22-06-23 Canada 35 Female Tablet 1 Month
2 3 Standard 12 28-02-23 27-06-23 United Kingdom 42 Male Smart TV 1 Month
3 4 Standard 12 10-07-22 26-06-23 Australia 51 Female Laptop 1 Month
4 5 Basic 10 01-05-23 28-06-23 Germany 33 Male Smartphone 1 Month
In [20]:
df['Join Date'] = pd.to_datetime(df['Join Date'], format='%d-%m-%y') 
df['Last Payment Date'] = pd.to_datetime(df['Last Payment Date'], format='%d-%m-%y')
In [24]:
df['Duration'] = (df['Last Payment Date'] - df['Join Date']).dt.days
In [27]:
df['Duration Months'] = df['Duration'].apply(lambda x: math.ceil(x/30))
In [28]:
average_duration = df['Duration Months'].mean()

print(f"Average Duration: {average_duration}")

Average Duration: 10.7808
In [29]:
df.head()
Out[29]:
User ID Subscription Type Monthly Revenue Join Date Last Payment Date Country Age Gender Device Plan Duration Duration Duration Months
0 1 Basic 10 2022-01-15 2023-06-10 United States 28 Male Smartphone 1 Month 511 18
1 2 Premium 15 2021-09-05 2023-06-22 Canada 35 Female Tablet 1 Month 655 22
2 3 Standard 12 2023-02-28 2023-06-27 United Kingdom 42 Male Smart TV 1 Month 119 4
3 4 Standard 12 2022-07-10 2023-06-26 Australia 51 Female Laptop 1 Month 351 12
4 5 Basic 10 2023-05-01 2023-06-28 Germany 33 Male Smartphone 1 Month 58 2
In [30]:
# Lifetime value

df['LTV'] = df['Monthly Revenue'] * df['Duration Months']
In [31]:
df.head()
Out[31]:
User ID Subscription Type Monthly Revenue Join Date Last Payment Date Country Age Gender Device Plan Duration Duration Duration Months LTV
0 1 Basic 10 2022-01-15 2023-06-10 United States 28 Male Smartphone 1 Month 511 18 180
1 2 Premium 15 2021-09-05 2023-06-22 Canada 35 Female Tablet 1 Month 655 22 330
2 3 Standard 12 2023-02-28 2023-06-27 United Kingdom 42 Male Smart TV 1 Month 119 4 48
3 4 Standard 12 2022-07-10 2023-06-26 Australia 51 Female Laptop 1 Month 351 12 144
4 5 Basic 10 2023-05-01 2023-06-28 Germany 33 Male Smartphone 1 Month 58 2 20

Avereage & Total LTV Customers¶

In [35]:
ltv_per_country_sum = df.groupby('Country')['LTV'].sum().sort_values(ascending=False)

plt.figure(figsize=(12,6))
ltv_per_country_sum.plot(kind='bar',color='brown')
plt.title('LTV per Country')
plt.xlabel('Country')
plt.ylabel(' LTV')
plt.xticks(rotation=45)
plt.show()
No description has been provided for this image
In [38]:
ltv_per_country = df.groupby('Country')['LTV'].mean().sort_values(ascending=False)

plt.figure(figsize=(12,6))
ltv_per_country.plot(kind='bar',color='brown')
plt.title('LTV per Country')
plt.xlabel('Country')
plt.ylabel(' LTV')
plt.xticks(rotation=45)
plt.show()
No description has been provided for this image

Revenue by Device¶

In [40]:
revenue_by_device = df.groupby('Device')['LTV'].sum()

plt.figure(figsize=(12,6))
revenue_by_device.plot(kind='bar',color='green')
plt.title('Revenue by Device')
plt.xlabel('Total Revenue')
plt.ylabel('Device')
plt.xticks(rotation=45)
plt.show()
No description has been provided for this image
In [41]:
plt.figure(figsize=(12,6))
sns.countplot(x='Device', hue='Gender', data=df)
plt.title('Gender distribution by Gender')
plt.xlabel('Device')
plt.ylabel('Count')
plt.xticks(rotation=45)
plt.show()
No description has been provided for this image

Metrics and Analyses¶

1.Number of Users per Country

The highest number of users are from the United States and Spain, each with 451 users, making up 18.04% of the total user base each. Following these are Canada with 317 users (12.68%), and the United Kingdom with 183 users (7.32%). Australia, Germany, France, Brazil, Mexico, and Italy each have the same number of users, 183, accounting for 7.32% of the total user base per country.

2.Gender Counts

The overall gender distribution among users is almost equal, with 1257 female users (50.28%) and 1243 male users (49.72%).

The gender distribution is almost equal across all countries, with a nearly balanced ratio of female and male users in each country. For example, in the United States, the distribution is 49.89% female and 50.11% male, and in Spain, it is 51.66% female and 48.34% male, reflecting a similar balance in other countries as well.

3.Device Counts

The distribution of devices among users is relatively balanced. The breakdown is as follows:

  • Laptops are used by 636 users (25.44%).
  • Tablets are used by 633 users (25.32%).
  • Smartphones are used by 621 users (24.84%).
  • Smart TVs are used by 610 users (24.40%).

4.Subscription Type Counts

The distribution of subscription types among users is as follows:

  • Basic: 999 users (39.96%)
  • Standard: 768 users (30.72%)
  • Premium: 733 users (29.32%)

5.Age Distribution

The age distribution of users shows the following statistics:

• Mean Age: 38.8 years

• Median Age: 39.0 years

• Minimum Age: 26 years

• Maximum Age: 51 years

• Standard Deviation: 7.17 years

• Most Common Ages:

• 30 years: 116 users

• 39 years: 116 users

6.Duration Between Join Date and Last Payment Date

The duration between the join date and the last payment date among users shows the following statistics:

• Mean Duration: 308.6 days

• Median Duration: 307.0 days

• Minimum Duration: 8 days

• Maximum Duration: 655 days

7.Lifetime Value (LTV)

The highest total LTV is in the United States, with an LTV of 60,445, accounting for 18.91% of the overall LTV. This is because the United States has the highest number of users. Similarly, Spain has a high total LTV of 59,269 (18.54% of the overall LTV) due to its large user base. Other countries, such as Canada and France, have total LTVs of 43,116 (13.48%) and 26,080 (8.16%) respectively. The total LTVs for the remaining countries, including Australia, Brazil, Germany, Italy, Mexico, and the United Kingdom, range between 23,618 and 25,990, each contributing between 7.38% and 7.84% to the overall LTV. The average Lifetime Value (LTV) per user varies across different countries. France has the highest average LTV at 142.51, followed closely by the United Kingdom at 142.02. Other countries with relatively high average LTVs include Australia (136.73), Brazil (136.50), and Canada (136.01). The United States, despite having the highest total LTV due to its large user base, has an average LTV of 134.02.

Countries like Germany (133.05), Mexico (132.39), and Spain (131.42) have average LTVs that are slightly lower, while Italy has the lowest average LTV at 129.06.

This analysis highlights that while the United States and Spain have high total LTVs due to their large user bases, countries like France and the United Kingdom lead in terms of average LTV per user.

Recommendations¶

1.Focus on High-LTV Countries:

• The United States and Spain have the highest total LTVs. Consider targeted marketing and retention strategies in these countries to maximize revenue.

• France and the United Kingdom, with the highest average LTVs, should also be prioritized for premium offerings and upsell opportunities.

2.Gender-Specific Campaigns:

• The gender distribution is nearly equal across all countries. Tailor marketing campaigns to address the preferences and needs of both male and female users equally.

• Utilize the balanced gender ratio to create inclusive content and promotions that appeal to a wide audience.

3.Device Optimization:

• With a relatively even distribution of devices used by subscribers (Laptops, Tablets, Smartphones, Smart TVs), ensure that the platform is optimized for all these devices.

• Develop device-specific features or promotions to enhance user experience and engagement.

4.Subscription Type Promotions:

• The Basic subscription type has the highest number of users. Consider introducing features or benefits to encourage upgrades to Standard or Premium plans.

• Implement targeted promotions to highlight the value of higher-tier subscriptions to Basic plan users.

5.Future Analysis:

• Conduct deeper analyses on user behavior, content preferences, and churn rates to refine marketing and retention strategies.

• Explore geographical trends and regional preferences to tailor content and offerings to different markets.

In [ ]: